It actually took a decade before the rumor became true... In 1988 I wrote out a brief outline, following Quillen's paper Higher algebraic K-theory I. It was overwhelming. I talked to Hy Bass, the author of the classic book Algebraic K-theory, about what would be involved in writing such a book. It was scary, because (in 1988) I didn't know even how to write a book. I needed a warm-up exercise, a practice book if you will.
The result, An introduction to homological algebra, took over five years to write.
By this time (1995), the K-theory landscape had changed, and with it my vision of what my K-theory book should be. Was it an obsolete idea? After all, the new developments in Motivic Cohomology were affecting our knowledge of the K-theory of fields and varieties. In addition, there was no easily accessible source for this new material. Nevertheless, I wrote early versions of Chapters I-IV during 1994-1999. The project became known as the ``K-book'' at this time.
In 1999, I was asked to turn a series of lectures by Voevodsky into a book. This project took over six years, in collaboration with Carlo Mazza and Vladimir Voevodsky. The result was the book Lecture Notes on Motivic Cohomology, published in 2006.
In 2004-2008, Chapters IV and V were completed. At the same time, the final steps in the proof of the Norm Residue Theorem were finished. (This settles not just the Bloch-Kato Conjecture, but also the Beilinson-Lichtenbaum Conjectures and Quillen-Lichtenbaum Conjectures.) The proof of this theorem is scattered over a dozen papers and preprints, and writing it spanned over a decade of work, mostly by Rost and Voevodsky. Didn't it make sense to put this house in order? It did. I am currently collaborating with Christian Haesemeyer in writing a self-contained proof of this theorem.